1

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

Transparency and accountability initiatives

– do they really make a difference?

September 2013

Introduction Transparency and accountability initiatives (TAIs) are increasingly promoted as effective ways to tackle government corruption and inefficiency.1 TAIs are based on the premise that increased information flows from government enable greater civil society oversight. This helps make governments more accountable to their citizens, leading to improvements in resource and revenue management and service delivery.2

Transparency and accountability has been high up on the international political agenda. It was a core element of the 2013 G8 Summit and continues to be a key part of other processes such as the Open Government Partnership, the Extractive Industries Transparency Initiative, Global Initiative for Fiscal Transparency (GIFT) and many more. But research analysing the benefits and challenges of TAIs continues to be relatively weak.

It is for this reason that in this short report we assess the benefits of TAIs, highlighting the factors that contribute to their successes in improving governance.3 We also note some of the criticisms levelled at them, and we conclude with recommendations for those wishing to strengthen these initiatives.

What are transparency and accountability initiatives? TAI strategies are sometimes referred to as ‘social accountability’ or ‘demand-side’ processes and are often citizen-led. They complement traditional forms of engagement between government and citizens, such as elections and bureaucratic procedures, and include:

● freedom of information (FOI) requests ● budget tracking ● natural resource payment and revenue monitoring ● aid transparency initiatives ● litigation proceedings

TAIs can also be ‘supply-driven’ – initiated and supported by the state – including, for example:

● FOI laws ● government multi-stakeholder initiatives such as the Extractive Industries Transparency

Initiative (EITI)

1 While the main focus of this paper is government accountability, TAIs may also apply to other duty bearers, such as multinational companies and international financial institutions. 2 R. McGee and J. Gaventa (2011) Shifting power? Assessing the impact of transparency and accountability initiatives, Institute of Development Studies (IDS), Working Paper 383 3 Cases are from the International Budget Partnership (IBP), Extractive Industries Transparency Initiative (EITI), Construction Sector Transparency Initiative (CoST) and other sources.

2

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

Definitions

Transparency: ‘Timely and reliable economic, social and political information which is accessible to all relevant stakeholders.’4

Transparency is essential to accountability, but insufficient on its own. Most TAIs recognise the equal importance of active citizen participation.5

Accountability: ‘Holding actors responsible for their actions.’6

Accountability involves both answerability (citizen monitoring of decisions made, with governments providing information and justification for actions) and enforceability (government responsiveness to pressure for change, and consequences for failing to fulfil obligations).

The impacts of TAIs Research suggests that the positive impacts of TAIs can be grouped into six categories: better service delivery; better budget utilisation and reduced leakages; greater state responsiveness; the building of democratic space for citizen engagement; empowerment of local voices; and an improved investment climate.7

1. Better service delivery

Improved delivery of services such as healthcare and education is one of the main aims of many TAIs. This goal may require the use of a range of TAIs, as the following example from Argentina shows. The Constitution of the City of Buenos Aires states that the government must guarantee and fund free public education for all children 45 days old and above. However, thousands of children – mostly from poorer neighbourhoods – did not have access to a school place. The NGO Civil Association for Equality and Justice (ACIJ) became aware of the problem and gathered evidence before issuing a report and producing a film to raise public awareness. They discovered that the relevant budget information was not available to the public, and they submitted FOI requests. When the government did not respond, a court action forced it to reveal the data. ACIJ’s subsequent analysis of this information brought to light an underspend of 35 per cent. The NGO filed a class-action suit, appealing to the courts to order the government to comply with its obligations. At the same time, they began out-of-court discussions with the government – helped by a new, more sympathetic Minister for Education. With the court’s oversight, a draft agreement was reached, which created new school places for children and prioritised the districts with highest need.8

4 D. Kaufmann and A. Bellver, ‘Transparenting transparency – initial empirics and policy applications’, presentation at the Pre-Conference on Institutional Change for Growth and Poverty Reduction in Low-Income Countries, International Monetary Fund, Washington DC, July 2005, p.4 5 Eg United Nations Development Programme (UNDP) (2010) Fostering social accountability: from principle to practice – a guidance note 6 M. Tisné, Transparency, participation and accountability: definitions, background note for Transparency and Accountability Initiative, quoted in R. McGee, J. Gaventa et al (2010) Synthesis report: review of impact and effectiveness of transparency and accountability initiatives, Transparency and Accountability Initiative and IDS, p.13 7 McGee and Gaventa (2011) Op. Cit., p.16 and, for example, D. Diamond, ‘Driving transparency in extractive industries to minimize risks for investors’, 11 March 2013, http://eiti.org/blog/driving-transparency-extractive-industries-minimize-risks-investors 8 F. Basch (2011) Children’s right to early education in the city of Buenos Aires: a case study on ACIJ’s class action, IBP

3

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

2. Better budget utilisation and reduced leakages

Once citizens have access to information, they can take action to ensure that budgets are correctly spent – and reduce the amount lost to corruption. In India, for example, the National Campaign for Dalit Human Rights (NCDHR) lobbied the government to hold it to its promise under the Scheduled Caste Sub-Plan (SCSP) to ensure that development spending benefited Dalits proportionally.9 The NCDHR, working with the Centre for Budget and Governance Accountability, carried out SCSP-related budget analysis and tracking from 2006. They filed a right-to-information petition and disseminated findings, including to MPs. They found evidence of funds being diverted to finance construction for the forthcoming 2010 Commonwealth Games in Delhi. The NCDHR published its evidence, which was subsequently discussed in parliament. This led the relevant minister to admit that the funds should not have been diverted and to promise to restore the money.

This story is repeated around the world. For example, in Nigeria, the Extractive Industries Transparency Initiative (EITI) uncovered US$2.6 billion of underpayments from 1999 to 2008 in taxes and royalties – more than the federal budget for health or education.10 The government has so far recovered US$443 million.11 Information disclosure has also enabled communities and institutions to prevent budget loss, saving them from lengthy and costly recovery procedures. For example, in Ethiopia the estimated costs of a road project had tripled between the feasibility and design stages. The disclosure of this information sparked media interest and led the Roads Authority and supervising engineer to agree to an alternative road plan. This saved approximately 40 million Birr (US$ 2.35 million) – equivalent to the cost of a new urban hospital.12

3. Greater state or institutional responsiveness

TAIs often call for specific action from government or public bodies that were previously unwilling to act. In this way, they may increase the likelihood of dialogue and cooperation with the state/institution over future projects or concerns. This was the experience of civil society organisation (CSO) HakiElimu in Tanzania. Tanzania had seen an impressive rise in school enrolment rates, but HakiElimu was concerned about the quality of education being provided.13 They established a partnership with the Teachers’ Union, and discovered that morale was low due to poor pay and working environments. More than three-quarters of teachers said their salary was inadequate, and most rural teachers complained of being paid late. HakiElimu ran a local and national campaign highlighting teachers’ poor working and living conditions. The campaign gained traction and the government reacted negatively, banning HakiElimu from working in schools. However, the resulting media and parliamentary attention simply increased awareness of, and support for, the campaign. Public debate continued until the government, faced with increasing criticism, met with HakiElimu and subsequently lifted the ban. HakiElimu then contacted the finance minister with a solution for paying salaries on time – a proposal which was implemented. Consequently, HakiElimu’s relationship with the government began to change. It began to consult with HakiElimu on addressing problems in education and subsequently instituted several reforms addressing issues raised by the CSO.

9 V. Ramachandran and S. Goel (2011) Tracking funds for India’s most deprived: the story of the National Campaign for Dalit Human Rights’ ‘Campaign 789’, IBP 10 EITI (2012) Nigeria EITI: making transparency count, uncovering billions 11 http://www.thisdaylive.com/articles/neiti-recovers-n69bn-unremitted-oil-revenue/133876/ 12 CoST (2011) Report on assurance process for Modjo-Arerti-Gobensa, Sembo-Gobensa and Metehbila-Metehara road project – Gindeber-Gobensa summary and MPI 13 IBP, Quality of education reforms: the case of HakiElimu’s campaign of 2005–2007 (no date)

4

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

4. Building new democratic spaces for citizen engagement

TAIs can enhance citizen engagement through participatory decision-making and multi-stakeholder initiatives. The city of Porto Alegre, Brazil, has seen major success in this area. It introduced participative budgeting, with citizens invited to consultations on the city’s budget allocations.14 In 1990, only 1,000 citizens were involved but, as confidence in participative budgeting grew, numbers increased to 26,000 by 2006. After participation began, spending appeared more concentrated in poorer areas; furthermore, there was a high completion rate for projects decided at participatory meetings. Sixty per cent of Porto Alegre’s citizens reportedly know how the budget is spent, and this TAI model has now spread to more than 70 cities in Brazil and other countries in Latin America and beyond.

5. Giving local voices the confidence to hold leaders to account

Communities may feel they do not have the information they need to hold public institutions to account. Or, for lack of information, they may be unaware that there is anything amiss. TAIs can address both scenarios, and empower people to hold public institutions to account. This was the case in India, where a group of labourers who had worked on famine relief sites had not been paid. The authorities said that the labourers had not worked, and cited the Official Secrets Act in order to withhold the information that would prove they had.15 The labourers staged a sit-in and, helped by the CSO, Mazdoor Kisan Shakti Sangathan (MKSS), they obtained the information under FoI laws. Subsequent village-level public hearings revealed many irregularities, and the authorities rapidly increased payments to the labourers.

TAIs can also increase the effectiveness of governance feedback loops, empowering communities to express their true opinions of those in power. For example, in Benin, a United Nations Development Programme (UNDP) project built the capacity of CSOs, the media and parliament to debate budget and poverty reduction policies. In the subsequent election, only 30 per cent of parliamentarians were re-elected. The UNDP believed this was because voters had gained confidence to choose only parliamentarians known to have worked in the public interest.16

6. Improved investment climate

Foreign investors take a risk in deciding to undertake projects in a new country: ‘[C]ertain companies operating in areas of political instability, or in countries suffering from violence, bribery and corruption, face business risks that can significantly affect their commercial prospects and thus returns to investors.’17 A country’s compliance with a TAI can therefore help to reassure foreign companies and investors that the government will observe recognised standards, thus enhancing business incentives to invest. The EITI claims that implementation of the initiative can create ‘an improved investment climate by providing a clear signal to investors and international financial institutions that the government is committed to greater transparency’.18

14 UNESCO, The experience of the participative budget in Porto Alegre, Brazil (no date) 15 R. Calland (2010) Review of impact and effectiveness of transparency and accountability initiatives: Annex 3, Freedom of information, IDS 16 UNDP (2010) Op. Cit. 17 J. McDowell, ‘EITI contributing to a better investment climate’, 7 September 2009, http://eiti.org/blog/eiti-contributing-better-investment-climate 18 EITI, ‘EITI benefits’, http://eiti.org/eiti/benefits

5

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

What contributes to TAIs’ success? Social and political change is complex. Many factors that affect TAIs’ impacts are interdependent, and every TAI is context-specific. Identifying key elements that affect their outcomes is therefore difficult. However, the above examples suggest that three main variables appear to determine the degree of success of any TAI: information accessibility and usefulness; civil society participation; and government or state responsiveness.

1. Information accessibility and usefulness

Not all data is information that will lead to change. It needs to be accessible to those who can use it, available at the right time, easy to comprehend and relevant to people’s lives.

Accessible information. As can be seen from the examples above, communities often find this is a required starting point for action, in order for them to engage with government. FOI laws now exist in more than 90 countries, meaning that CSOs can still obtain information even if governments have not proactively released it.19 Once citizens have this data, they are then ‘armed’ to take action if they wish. For example, in Mexico, a suspected bias in the distribution of agricultural subsidies led the Subsidios al Campo campaign to use FoI requests and appeals to obtain and publish online data about subsidy payments and beneficiaries.20 The campaign discovered that, among other irregularities, 557 Ministry of Agriculture officials had been receiving millions of pesos in subsidies illegally. Their subsequent campaign, and the proof they had collected via their information-gathering, led to the state’s supreme audit institution ordering all officials to return the money wrongly paid to them.

Well-timed information. Having information that is up-to-date and on time is a particularly important factor where TAIs relate to budget transparency.21 A delay in the release of spending information impedes scrutiny by audit institutions and civil society, and weakens the impact the TAI can have. Such delays can also reduce the effectiveness of other TAIs. For example, in Peru, EITI data for 2004–2007 was published only in 2010, by which time it was of limited use.22 The EITI Board has sought to address this, and the EITI standard now requires reports to cover data ‘no older than the second to last complete accounting period’.23 However, the issue is still a concern particularly in relation to non-EITI TAIs.

Information that is easy to understand. If availability of information is a prerequisite, then ensuring that information is comprehensible is equally important. Information may need to be researched, interpreted and re-presented so that end users can understand it and therefore take action. The importance of information being comprehensible has been highlighted by the UK Public Accounts Committee. It warns: ‘Simply dumping data without appropriate interpretation can be of limited use… There is a clear benefit to the public when government data is analysed and interpreted by third

19 Right2Info, ‘Access to information laws: overview and statutory goals’, http://www.right2info.org/access-to-information-laws/access-to-information-laws-overview-and-statutory 20 G.M. Cejudo (2012) Evidence for change: the case of Subsidios al Campo in Mexico, IBP 21 IBP, ‘Rankings & key findings’, http://internationalbudget.org/what-we-do/open-budget-survey/rankings-key-findings/ 22 Tearfund, Revenue Watch and Paz y Esperanza (2012) Dissemination and impact of EITI national reports in Peru: summary report 23 EITI (2013) The EITI standard, p.15

6

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

parties.’24 The International Chamber of Commerce makes the same point: ‘The underlying challenge is to ensure that civil society can utilise information disclosed in a meaningful way.’25

This point is well illustrated by the following successful campaign in India. The National Campaign for Dalit Human Rights (NCDHR) discovered that not all government departments were coding spending earmarked for Dalits, making it extremely difficult to work out if funds were being used legitimately. The NCDHR took a systematic approach to explaining this problem to activists, organising workshops on coding, budget analysis and tracking earmarked funds. Its clear aim was to make the information it had gathered comprehensible to the general public. This work mobilised large numbers of demonstrators who attended a rally culminating in a day-long sit-in. The protest led to the government issuing a deadline by which date all departments must start coding their Dalit spending. This enabled CSOs to track budgets, which resulted in the return of misused funds. None of this would have been possible without NCDHR making the information it had gathered clear and easy to understand.26

Relevant information. Booth’s analysis indicates that people will participate more readily in governance issues if there is a clear benefit to them personally, such as improved education, agricultural inputs or roads.27 This was the case in Uganda, where parents were heavily involved in reducing the leakage of government funds for primary school fees. The schools relied primarily on parental contributions to cover non-wage expenditure, such as maintenance and school materials. This created a powerful incentive for parents to demand financial information from the schools and thus meant that individual schools had fairly good accounting systems. When Public Expenditure Tracking Surveys (PETS) were established, this information enabled researchers to see that there was an enormous disparity between what the government was funding and what schools were receiving. A public information campaign followed, and leakage was reduced from 74 per cent to less than 20 per cent in five years. It was found that leakage reduced significantly more in schools closer to the nearest newspaper vendors, which researchers took as proof that the availability of relevant information was key to campaign success.28

2. Civil society participation

As can be seen from the above examples, information alone is not enough to bring about change: citizens must then use it to demand answers. But what motivates people to do so?

According to Hirschman, when an individual is concerned about a public service, he or she has three possible ways to respond: ‘exit’ – to cease using the service, preferably replacing it with another (rarely an option for the poorest); ‘voice’ – to speak out about concerns; or ‘loyalty’ – to remain loyal despite having grievances.29 Without a viable alternative or democratic space for accountability, poor

24 House of Commons Public Accounts Committee (2012) Implementing the transparency agenda. http://www.publications.parliament.uk/pa/cm201213/cmselect/cmpubacc/102/10204.htm, paragraph 2 25 International Chamber of Commerce UK (2012) EU Transparency Directive, http://www.international-chamber.co.uk/blog/2012/04/26/eu-transparency-directive/ 26 V. Ramachandran and S. Goel (2011) Op. Cit. 27 D. Booth (2012) Development as a collective action problem: addressing the real challenges of African governance – synthesis report, Africa Power and Politics Programme and Overseas Development Institute 28G. Sundet (2008) Following the money: do public expenditure tracking surveys matter? U4 Anti-Corruption Resource Centre 29 A.O. Hirschman (1970) Exit, voice, and loyalty: responses to decline in firms, organizations, and states, Harvard University Press

7

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

people may be trapped in ‘loyalty’.30 However, positive experiences and a tradition of engagement and successful TAIs are likely to help them participate, speak out and use their ‘voice’.

Key factors that influence the degree to which citizens participate in TAIs include: capacity and confidence; relationships of trust and belief that change is possible; incentives; civil society links to wider networks; and media involvement.

Capacity and confidence. Effective TAIs can build capacity and confidence in citizens, empowering them to take action when otherwise they might not have done. An example from Pakistan is a case in point. The 2005 earthquake in Hazara destroyed housing and local services. The government established the Earthquake Reconstruction Rehabilitation Authority (ERRA) to oversee reconstruction, and within two years it claimed to have reached 99 per cent of eligible people with 600,000 housing grants. This claim did not appear to be reflected in the situation on the ground, and the Oman Asghar Khan Development Foundation conducted research to investigate this claim. The foundation also discovered that only 175 schools and 38 health units had been rebuilt, against respective targets of 5,344 and 307. It built the capacity of affected citizens to understand what had happened and to take action, overcoming the constraints of the clan culture to agree together what they wanted to campaign for. It then helped give citizens the confidence to act. As one protestor noted, many people had been living in a state of ‘helplessness and hopelessness’ up to this point. Citizens then took a number of actions, ranging from media work to public protests/sit-ins. The result was that the government accelerated the pace of reconstruction significantly. Completion rates for health, water/sanitation and education projects improved markedly. Rural housing subsidies went from reaching 39.1 per cent of those eligible to 81.2 per cent – in just one year.31

Trust and belief that change is possible. Countries with a tradition of citizen engagement and protest, such as India, have shown very positive results from TAIs. With a culture of accountability, people expect state institutions to respond, or even to consult proactively.32 And citizens are ‘more likely to be involved in monitoring the implementation of government programmes if they have also been involved in shaping them’.33 This was the experience in Porto Alegre, Brazil, where citizens were invited to consultations on the city’s budget allocations (see page 4).34 Participation increased year on year as the municipal government responded to citizens’ demands. But when the city’s ruling party changed and the new incumbents placed less emphasis on local hearings, participation decreased.

Incentives. As noted above, citizens are more likely to engage with the state on issues that have a direct impact on their lives. For Tembo, a key factor is therefore ‘finding the right incentives to mobilise and aggregate citizens’ voices while also generating incentives for engaging state institutions’.35 This was found to be the case during research into the EITI in Peru, where most of those interviewed wanted to know how industry payments contributed to development in their

30 D.W. Walker, ‘Citizen-driven reform of local-level basic services: community-based performance monitoring’, Development in Practice, 19.8, 2009, pp. 1035–51 31 IBP, Earthquake reconstruction in Pakistan: the case of the Omar Asghar Khan Development Foundation’s campaign (no date) 32 P. Newell and J. Wheeler (2006) Making accountability count, IDS, Policy Briefing 33 33 P. Houtzager, A. Joshi and A. Gurza Lavalle (eds), ‘State reform and social accountability’, IDS Bulletin, 38.6, 2008, cited in McGee and Gaventa (2011) Op. Cit. 34 UNESCO, The experience of the participative budget in Porto Alegre, Brazil (no date) 35 F. Tembo (2012) Citizen voice and state accountability: towards theories of change that embrace contextual dynamics, ODI, Project Briefing 73

8

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

communities – not the amounts paid to central government.36 It should be noted then, that in resource-dependent countries, ‘providing highly aggregate macroeconomic figures on oil revenues or expenditures is likely to provide only weak individual incentives’, and is therefore unlikely to prompt citizens to engage in governance concerns.37

Links between civil society and broad-based coalitions. TAIs which involve CSOs linking with other actors from a range of interest groups in coalition can be more successful in turning up the pressure on governments and institutions to make changes. Links to trade unions, churches, international CSOs, donors and international financial institutions give ‘safety in numbers’, encourage citizen participation and strengthen campaigns. For example, the coalition campaigning for justice and equality in Mexican agricultural subsidies (page 5) comprised a peasant organisation, a Mexico City thinktank and a US partner skilled in visualising complex datasets. In isolation, individual groups could not have brought change: the IT skills, policy expertise and grassroots knowledge were all essential.

Media involvement. The media is a powerful tool in TAIs when it is used to educate citizens about rights, build debate, break stories, question the status quo and shame underperforming officials. Naurin and Lindstedt consider strong media coverage to be key in preventing corruption.38 For example, in Benin, UNDP increased the capacity of the Beninese press to engage in and scrutinise budget setting, including how to analyse parliamentary discussions (including parliamentary question times). This resulted in increased media coverage of the national budget and poverty reduction strategy. This in turn contributed to better understanding of the issues among the public, and helped to ensure greater scrutiny of the use of public resources.39

Brunetti and Weder noted press freedom was a strong weapon against corruption.40 Suphachalasai found the same to be true for media competition.41 For Lindstedt and Naurin, press freedom is more effective against corruption when it combines with improved education.42

3. Government or state responsiveness

In this section, we look at the factors that contribute to government or state responsiveness. McGee and Gaventa discuss accountability as ‘ideally involv[ing] both answerability – the responsibility of duty-bearers to provide information and justification about their actions – and enforceability – the possibility of penalties or consequences’ for their failure to respond meaningfully.43 The UK Department for International Development (DFID) has drawn attention to the importance of government commitment to inform and engage at all levels.44 A number of factors contribute to – and

36 Tearfund, Revenue Watch and Paz y Esperanza (2012) Op. Cit. 37 I. Kolstad, A. Wiig and A. Williams (2008) Tackling corruption in oil rich countries: the role of transparency, U4 Anti-Corruption Resource Centre, p.3 38 C. Lindstedt and D. Naurin (2005) Transparency against corruption: a cross country analysis, cited in H. Oscarsson (2008) Media and quality of government: a research overview, University of Gothenburg, QoG Working Paper Series 12 39 UNDP (2010) Op. Cit. 40 A. Brunetti and B. Weder, ‘A free press is bad news for corruption’, Journal of Public Economics, 87.7-8, 2003, pp.1801–24 41 S. Suphachalasai (2005) Bureaucratic corruption and mass media, University of Cambridge, Department of Land Economics, Environmental Economy and Policy Research Working Paper 5 42 Lindstedt and Naurin (2005) Op. Cit. 43 McGee and Gaventa (2011) Op. Cit., p.13, citing A.M. Goetz and R. Jenkins (2005) Reinventing accountability: making democracy work for human development, Palgrave Macmillan 44 DFID, Public expenditure tracking surveys (no date) http://webarchive.nationalarchives.gov.uk/+/http:/www.dfid.gov.uk/aboutdfid/organisation/pfma/pfma-pets.pdf

9

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

can improve – responsiveness: political context and political will, government capacity, horizontal accountability, and external pressure.

Political context and political will. Political will is difficult to influence and, without it, the options for TAIs are severely limited. However, where governments are willing (and able) to act, there can be a crucial window of opportunity for TAIs.45 Hubbard found that the success of the PETS in Uganda example above (page 6) was due as much to government reforms as to the campaign itself.46 This conclusion is perhaps confirmed by the example of neighbouring Tanzania, where PETS were also introduced, but did not produce the major decrease in leakage found in Uganda.

Government capacity. Governments can, of course, be willing to act – but are sometimes unable to do so. This too can ultimately impact the success of TAIs. For example, central government policy decisions can prove difficult for local authorities to implement if they lack institutional capacity. The CSO Samarthan found this to be the case in India. The National Rural Employment Guarantee Act promises people in poor rural communities 100 days’ employment per year in unskilled occupations.47 The ruling party were pushing for implementation, but work was not being offered to communities. In part, this was because citizens were unaware they were entitled to this work – or were being intimidated by officials and scared off. Samarthan therefore educated communities on their rights, and accompanied them as they sought work. However, Samarthan also discovered that part of the problem was that the village-level functionaries tasked with offering work lacked the training they needed to be effective. Samarthan therefore offered capacity-building technical training on planning, budgeting and administration, which helped ensure that work projects were made available. Without this training (which was not being provided by the government), local officials would simply have been unable to deliver, no matter how much demand there had been from citizens.

Horizontal accountability. Civil society can obtain the information they need and campaign for change but they cannot force an institution or the state to change. As seen in the Argentinian and Mexican examples above (on pages 2 and 5), citizens may need the support of legislative, audit and judicial institutions that oversee government bodies and provide redress for wrongdoing, if change is to be enforced. Fukuda-Parr et al also emphasise this point, noting that effective functioning of the legislative and audit institutions is crucial for the budget development process.48

External pressure. Fox’s discussion of the state–citizen relationship suggests that there are three sources of pressure for change: from inside the agency (whistle-blowers and pro-reform allies); from outside the agency and inside the state (CSOs); and from outside the state (donors, international financial institutions and private investors).49 External peer-to-peer pressure is a key element in the EITI, assuming that greater transparency will reduce investor risk and increase investment. Supranational bodies such as the World Bank and International Monetary Fund can encourage countries to be more transparent and accountable, using praise or threat of sanctions. External peer-to-peer pressure is also a central idea behind the Open Government Partnership, a voluntary initiative

45 A. Rocha Menocal and B. Sharma (2008) Joint evaluation of Citizens’ Voice and Accountability: synthesis report, DFID 46 P. Hubbard (2007) Putting the power of transparency in context: information’s role in reducing corruption in Uganda’s education sector, Center for Global Development, Working Paper 136 47 R. Awasthi (2011) Samarthan’s campaign to improve access to the National Rural Employment Guarantee Scheme in India, IBP 48 S. Fukuda-Parr, P. Guyer and T. Lawson-Remer (2011) Does budget transparency lead to stronger human development outcomes and commitments to economic and social rights?, IBP, Working Paper 4, p.19 49 J. Fox (2007) Accountability politics: power and voice in rural Mexico, Oxford University Press, cited in McGee, Gaventa et al (2010) Op. Cit.

10

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

where states commit to become more open. This mechanism encourages each state to make progress on open government, allows space for external pressure and helps creates a ‘race to the top’.

Critiquing TAIs This paper can offer only a rough sketch of the common criticisms of TAIs. The review of TAIs commissioned by DFID investigates and addresses some of these concerns.50 Charges include claims that:

● There is insufficient empirical evidence of TAI success.51 ● TAIs make misplaced assumptions about latent civil society demand waiting to hold

governments to account for the greater good.52 ● TAIs are mechanistic and neglect issues of power and politics.53 ● TAIs are focused on immediate outcomes more than on wider governance concerns.

The Africa Power and Politics Programme has levelled a more fundamental charge at the very framing of TAIs, arguing that ‘governance challenges are not fundamentally about one set of people getting another set of people to behave better... [but] about both sets of people finding ways of being able to act collectively in their own best interests’ in fragmented societies.54 It is therefore important that TAIs address the complexity of governance issues and look to develop understanding and relationships between the different groups involved.

Conclusions As this paper has shown, there is considerable evidence from across the world showing how TAIs have led to better service delivery, better budget utilisation and reduced leakages, greater state responsiveness, the building of democratic space for citizen engagement, empowerment of local voices, and an improved investment climate. Communities have seen their local services improve and have witnessed increased investment in their children’s futures, fairer distribution of grants and a decrease in the likelihood of corruption. These successes have been hard won: CSOs have had to fight with determination in the face of many setbacks, stand firm in the face of government intimidation and spend years gathering the evidence they needed.

We have seen that there are some critical factors which can contribute to the success or otherwise of these efforts. Information at the right time, for the right place, available in a format people can understand, is often an absolutely foundational starting point. However, to turn that information into action, civil society must feel motivated to act upon it. CSOs may need to build both their own, and citizens’, capacity to do so, and build citizens’ confidence to believe that change can happen. Citizen and CSO efforts can be significantly strengthened by being linked in with networks, and by media involvement. Ultimately, this can all be to no avail in the face of a state or institution that simply does not want to change. The success of TAIs does require the state to be responsive to CSOs and citizens concerns. Civil society therefore can and must seek to change this – although it is difficult – by using a robust combination of external pressure and horizontal accountability.

50 R. McGee, J. Gaventa et al (2010) Op. Cit. 51 D. Booth ( 2012) Op. Cit. 52 Ibid 53 McGee and Gaventa (2011) Op. Cit. 54 D. Booth ( 2012) Op. Cit., pp viii, xi

11

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

Policy recommendations for those working to strengthen TAIs

a. Ensure there is access to relevant, timely and reliable information. Governments should make available high-quality information, in sufficient time for civil society to scrutinise it. If they do not do so, CSOs should call for it to be provided. Once the information is made available, ‘infomediaries’ (researchers, the media, CSOs etc) need to be supported to make independent analysis of the data and make the information useable and comprehensible to the public, to open up the issues to public debate.

b. Address both supply and demand dynamics. Without demand from CSOs and citizens, there is little incentive for governments to open up budgets and policy-making. Similarly, demand without supply will lead to frustration and increased tension between government and civil society. Neither dimension alone will bring change.55

c. Public participation can be important but is not always essential. CSOs need to analyse the situation and identify the method most likely to achieve change. Public mobilisation is not the only route to accountability; litigation and media exposure are also powerful means to prompt a government response. Where public participation is needed/helpful, CSOs should identify the best method for engaging the public – sometimes a persuasive incentive is needed. Or they may need to reassure citizens that their voices will be heard.

d. Seek to address government and political responsiveness. CSOs should build relationships between the citizens involved in TAIs and governments, so that both see the benefits of accountability. In some cases, the institutional capacity of the government needs to be built so that it has the capacity to be responsive. In other cases, a lack of will on behalf of the state to respond to citizens’ concerns needs to be addressed. Governments need to be responsive to citizens’ concerns – even if they cannot provide the service/outcome that the TAI has called for; they should still provide an explanation as to why they cannot act. Deepened relationships and understanding between both parties can help to address this through building trust and goodwill.

e. Seek to build alliances and relationships, considering unusual partnerships. TAIs often involve a markedly political process and governments may put obstacles in the way of accountability. In these cases, CSOs should build alliances with pro-reform actors within government or pro-reform international interests, and seek to strengthen horizontal accountability mechanisms. CSOs should seek alliances across specialties, to gather all the skills required. The more pressure that a TAI can bring from every direction, the more likely a government is to respond constructively.56

f. Ensure media independence. A free media, independent of government and of sectional interests, is uniquely positioned to publish in the public interest domestically and internationally, educating citizens and external actors in ways that help exert pressure on governments to respond. Governments should therefore ensure that the media has freedom to act and, where governments do not do so, CSOs should call for media independence.

55 T. O’Neil, M. Foresti and A. Hudson (2007) Evaluation of citizens’ voice and accountability: review of the literature and donor approaches report, ODI; D. Booth (2012) Op. Cit.; International Chamber of Commerce UK (2012) Op. Cit. 56 J. Fox (2007) Op. Cit., cited in McGee , Gaventa et al (2010) Op. Cit.

12

Registered Charity No. 265464 (England and Wales) SC037624

www.tearfund.org

This short report is part of Tearfund’s ongoing work on governance and corruption. Joint partner-led reports in Peru, Zambia and Tanzania support the findings above and look at the impacts of transparency and participation on service delivery in these three contexts.57 Opportunities for future research include a cost-benefit analysis of TAIs and the impact of TAIs on investor choices.

For further information, please contact:

Melissa Lawson: melissa.lawson@tearfund.org

Catherine Candish: Catherine.candish@tearfund.org

Tearfund is a Christian relief and development agency building a global network of local churches to help eradicate poverty. Tearfund publications are available at www.tearfund.org/tilz

Registered Charity No. 265464 (England and Wales) SC037624 (Scotland)

www.tearfund.org

57 Tearfund (2013) Peru: dissemination and impact of EITI national reports in Peru;

Tearfund (2013) Zambia: Constituency Development Funds – transparency in grassroots development or political patronage;

Tearfund (2013) Tanzania: following the money – a quest for social accountability in Tanzania